[owc] Weekly src changes ending 2007-01-21

OWC auto at squish.net
Mon Jan 22 07:02:53 GMT 2007


OpenBSD src changes summary for 2007-01-14 to 2007-01-21 inclusive
==================================================================

distrib/sets                            etc/etc.landisk/sysctl.conf
gnu/usr.bin/gcc                         lib/libc
sbin/dhclient                           sbin/pfctl
sbin/route                              share/man
sys/adosfs                              sys/arch/amd64/amd64
sys/arch/amd64/isa                      sys/arch/amd64/pci
sys/arch/hp300/dev                      sys/arch/hppa/hppa
sys/arch/hppa/include                   sys/arch/i386/i386
sys/arch/i386/pci                       sys/arch/landisk/conf
sys/arch/landisk/dev                    sys/arch/landisk/landisk
sys/arch/luna88k/conf                   sys/arch/mac68k/dev
sys/arch/macppc/dev                     sys/arch/macppc/macppc
sys/arch/sh/include                     sys/arch/sh/sh
sys/arch/sparc64/dev                    sys/arch/sparc64/sparc64
sys/arch/sparc64/stand/ofwboot          sys/arch/vax/vax
sys/compat/netbsd                       sys/dev/acpi
sys/dev/ic                              sys/dev/mii
sys/dev/pci                             sys/dev/usb
sys/isofs/cd9660                        sys/kern
sys/miscfs/fifofs                       sys/miscfs/specfs
sys/msdosfs                             sys/net
sys/netinet                             sys/netinet6
sys/nfs                                 sys/scsi
sys/sys                                 sys/ufs/ext2fs
sys/ufs/ffs                             sys/ufs/ufs
sys/uvm                                 sys/xfs
usr.bin/chpass                          usr.bin/compress
usr.bin/cvs                             usr.bin/lex
usr.bin/make                            usr.bin/rcs
usr.bin/ssh                             usr.bin/yacc
usr.sbin/hoststated                     usr.sbin/ntpd
usr.sbin/ospfd                          usr.sbin/ripd
usr.sbin/vnconfig                       

== distrib =========================================================== 01/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/distrib

sets

  ~ lists/comp/mi                         ~ lists/man/mi

  > sync (pvalchev@)

== etc =============================================================== 02/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/etc

etc.landisk/sysctl.conf

  ~ etc.landisk/sysctl.conf               

  > power(4) driver for the power switch on many landisk models, hooked
  > up to machdep.kbdreset; modelled after the sparc64 power(4) driver
  > discussed with miod@ and jsg@ (martin@)

== gnu =============================================================== 03/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/gnu

usr.bin/gcc

  ~ gcc/config.gcc                        

  > Remove references to xm files that no longer exist.  Taken from GCC 3.4.
  > ok brad@ (tsi@)

  ~ gcc/config/alpha/alpha.c              

  > -Wstack-larger-than support, copied from gcc2. (miod@)

== lib =============================================================== 04/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/lib

libc

  ~ stdlib/imaxabs.3                      

  > sync with the code: int -> intmax_t
  > ok jmc@ (steven@)

  ~ stdio/vfprintf.c                      

  > Fix potential int overflow for printf(3) when passing in very large
  > values for the field width.  Adapted from a diff by Christian Biere.
  > (millert@)

== sbin ============================================================== 05/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/sbin

dhclient

  ~ options.c                             

  > Don't treat any packet with an invalid option buffer or an invalid
  > message type option of 0 as a BOOTP offer. Don't process any option
  > buffer after one is found to be corrupt. Don't process overflow
  > buffers in a BOOTP packet.	At least try to accept DHCP offers even if
  > a bad options buffer is encountered.
  > Brings code more into line with current ISC dhclient.
  > ok stevesk@ (krw@)

  ~ dhclient.c                            ~ dhcpd.h
  ~ options.c                             

  > More cleanup. Eliminate 'struct packet' by cleverly passing as
  > parameters just those things actually needed and using existing
  > structs and information. No functional change.
  > ok stevesk@ (krw@)

pfctl

  ~ pfctl.8                               ~ pfctl.c
  ~ pfctl_table.c                         

  > implement -T expire.
  > "pfctl -t tablename -T expire 3600" would expire all entries in the given
  > table that are older than 3600 seconds. ok dhartmei, manpage help & ok jmc
  > (henning@)

route

  ~ route.8                               

  > reference ripd(8) vs. routed(8) and remove routed statement in BUGS
  > that no longer applies because it is not referenced.  ok claudio@ jmc@
  > (stevesk@)

  ~ route.8                               

  > add an EXAMPLES section;
  > from mark lumsden via otto, tidied up somewhat by myself;
  > ok otto claudio (jmc@)

== share ============================================================= 06/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/share

man

  ~ man4/zyd.4                            

  > Match some additional devices found in Linux driver. (jsg@)

  ~ man4/re.4                             

  > Update to match reality.  From brad. (weingart@)

  + man4/man4.landisk/Makefile            + man4/man4.landisk/power.4

  > document what one can do with the power switch (martin@)

  ~ man4/Makefile                         

  > descend into man4.landisk (martin@)

  ~ man9/uvm.9                            

  > Implement multiple segment allocation for uvm_pglistalloc, which fixes
  > most agp_generic_bind_memory failures when memory is limited and very
  > fragmented.
  > In effect, this should fix a lot of X startup crashes after
  > activities that exercise memory a lot (e.g. make builds, building big
  > ports, etc).
  > ok mickey, miod (dim@)

  ~ man9/VOP_LOOKUP.9                     

  > Remove commented out references to VOP_LEASE();
  > ok jmc@ (who pointed this out to me) (thib@)

  ~ man5/bsd.port.mk.5                    

  > describe BULK_DO variable (steven@)

  ~ man4/acpitz.4                         ~ man4/acpicpu.4
  ~ man4/acpiac.4                         

  > Somehow the .Xr acpi 4 fell out of the SEE ALSO section, so put it back.
  > (mk@)

  ~ man4/axe.4                            

  > Add support for AX88772 devices with external PHY.
  > Thanks to Marco S Hyman for sending me hardware to play with.
  > Committed over DUB-E100 rev B1 with external IC Plus PHY. (jsg@)

  ~ man5/bsd.port.mk.5                    

  > FULLPKGPATH is no longer read-only, and COMMENT is also subpackage
  > dependent.
  > ok sturm@ bernd@ (steven@)

  ~ man4/usb.4                            ~ man4/udav.4
  ~ man4/url.4                            

  > add `10/100' to .Nd; from brad (jmc@)

== sys =============================================================== 07/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/sys

adosfs

  ~ advnops.c                             

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

arch/amd64/amd64

  ~ Locore.c                              ~ apic.c
  ~ autoconf.c                            ~ bus_dma.c
  ~ bus_space.c                           ~ conf.c
  ~ cpu.c                                 ~ db_disasm.c
  ~ db_interface.c                        ~ db_trace.c
  ~ disksubr.c                            ~ gdt.c
  ~ intr.c                                ~ ioapic.c
  ~ lapic.c                               ~ machdep.c
  ~ mem.c                                 ~ mpbios.c
  ~ pmap.c                                ~ rbus_machdep.c
  ~ sys_machdep.c                         ~ syscall.c
  ~ trap.c                                

  > ansi/deregister (jsg@)

arch/amd64/isa

  ~ clock.c                               ~ isa_machdep.c

  > ansi/deregister (jsg@)

arch/amd64/pci

  ~ aapic.c                               ~ pchb.c
  ~ pci_machdep.c                         ~ pcib.c
  ~ pciide_machdep.c                      

  > ansi/deregister (jsg@)

arch/hp300/dev

  ~ frodo.c                               ~ isabr.c

  > Move the isa bridge probe from isabr to frodo, so that frodo only attempts
  > to attach isabr if it really exists. This gets rid of ``isabr0 not
  > configured''
  > messages on 425e. (miod@)

arch/hppa/hppa

  ~ trap.c                                

  > undo last broken userret() change here; miod@ ok (mickey@)

  ~ trap.c                                

  > new shorter version of userret() as surely miod ment it to be like (; miod@
  > ok (mickey@)

arch/hppa/include

  ~ cpu.h                                 

  > new shorter version of userret() as surely miod ment it to be like (; miod@
  > ok (mickey@)

arch/i386/i386

  ~ bios.c                                

  > No longer print the 'bios id' field in the dmesg, its of dubious diagnostic
  > value and changes with acpi versus non acpi kernels.
  > ok deraadt@ tom@ (gwk@)

arch/i386/pci

  ~ glxsb.c                               

  > Update comment (on how long we wait for operation to complete) reflect
  > reality, and remove the maxpolls stuff at the same time.  Still disabled
  > in-tree. (tom@)

arch/landisk/conf

  ~ GENERIC                               ~ files.landisk

  > power(4) driver for the power switch on many landisk models, hooked
  > up to machdep.kbdreset; modelled after the sparc64 power(4) driver
  > discussed with miod@ and jsg@ (martin@)

arch/landisk/dev

  + power.c                               

  > power(4) driver for the power switch on many landisk models, hooked
  > up to machdep.kbdreset; modelled after the sparc64 power(4) driver
  > discussed with miod@ and jsg@ (martin@)

arch/landisk/landisk

  ~ landiskreg.h                          

  > power(4) driver for the power switch on many landisk models, hooked
  > up to machdep.kbdreset; modelled after the sparc64 power(4) driver
  > discussed with miod@ and jsg@ (martin@)

arch/luna88k/conf

  ~ RAMDISK                               

  > Same as GENERIC, m88k_cpus[] should have 4 elements on luna88k, even in
  > non-MULTIPROCESSOR kernel.
  > ok miod@ (aoyama@)

arch/mac68k/dev

  ~ zs.c                                  

  > zsprintf() is gone, fixes ZSMACDEBUG case (martin@)

arch/macppc/dev

  ~ zs.c                                  

  > zsprintf() is gone, fixes ZSMACDEBUG case (martin@)

arch/macppc/macppc

  ~ locore.S                              

  > Remove some cruft, SUPPORT_PPC1BUG isnt defined or needed. While here fix
  > some whitespace nits. No binary change.
  > ok drahn@ (gwk@)

  ~ locore.S                              

  > Use addic. %r31,%r31,1 instead of addic. %r31,%r31,%r1. (kettenis@)

arch/sh/include

  ~ cpu.h                                 

  > power(4) driver for the power switch on many landisk models, hooked
  > up to machdep.kbdreset; modelled after the sparc64 power(4) driver
  > discussed with miod@ and jsg@ (martin@)

arch/sh/sh

  ~ sh_machdep.c                          

  > power(4) driver for the power switch on many landisk models, hooked
  > up to machdep.kbdreset; modelled after the sparc64 power(4) driver
  > discussed with miod@ and jsg@ (martin@)

arch/sparc64/dev

  ~ schizoreg.h                           

  > Add another Tomatillo-specific register.
  > Fix definition of TOM_PCICTRL_DTO_INT. (kettenis@)

  ~ schizovar.h                           

  > Add macros to access Schizo's PCI config space. (kettenis@)

  ~ schizo.c                              

  > Provide a bit more diagnostic information in the PCI error handler.
  > (kettenis@)

  ~ schizo.c                              

  > Clear error bits in PCI command/status register before establishing PCI
  > error
  > handler. (kettenis@)

  ~ schizo.c                              

  > Really print iommu TFAR (instead of CTRL again). (kettenis@)

  ~ schizo.c                              ~ schizoreg.h

  > Do not panic on non-fatal iommu errors. (kettenis@)

arch/sparc64/sparc64

  ~ ofw_machdep.c                         ~ openfirm.c

  > Remove unnecessary ampersands in ADR2CELL macro invocations;
  > Change OF_printf references to printf, given the former doesn't exist;
  > Clean up formatting, whitespace, unused code, etc.
  > ok miod@, kettenis@ (tsi@)

arch/sparc64/stand/ofwboot

  ~ Locore.c                              

  > Remove unnecessary ampersands in ADR2CELL macro invocations;
  > Change OF_printf references to printf, given the former doesn't exist;
  > Clean up formatting, whitespace, unused code, etc.
  > ok miod@, kettenis@ (tsi@)

arch/vax/vax

  ~ random.s                              

  > ragge at netbsdxed the copyright, thanks. ok miod@ (otto@)

compat/netbsd

  ~ netbsd_misc.c                         

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

dev/acpi

  ~ dsdt.c                                

  > Fix invalid _CRS length.  Some aml implementations lie about how long a
  > _CRS buffer is and this compensates for it by fixing up the minimum value.
  > This fixes several machines that end up with invalid irq/ioapic values.
  > KNF nit as well
  > Code from jordan, tested on Intel SHG2 board (he committed this agains my
  > local tree) (marco@)

  ~ dsdt.c                                

  > Added fix for static integer
  > ok marco@ (jordan@)

  ~ acpidev.h                             ~ acpiprt.c

  > Check whether PCI interrupt link devices are enabled.
  > Based on a diff from marco at .ettenis@)

dev/ic

  ~ bcw.c                                 

  > Introduce register read/write helper routines for radio and phy.  Start
  > to add bcw_radio_on() and bcw_radio_channel().
  > ok Jon Simola (mglocker@)

  ~ ciss.c                                

  > do not freak out if cannot map physdrives into volumes and handle it proper
  > everywhere (mickey@)

  ~ elink3.c                              

  > Some '= 0' -> '= NULL' cleanup for pointer fields and return values.
  > From Brad.
  > Tested by nick@ just to be sure. (krw@)

dev/mii

  ~ rgephy.c                              

  > Use MII_ANEGTICKS_GIGE instead of MII_ANEGTICKS. (kettenis@)

dev/pci

  ~ if_em.c                               

  > Report flow control status.
  > From brad@ (kettenis@)

  ~ if_msk.c                              

  > Remove some textual sk heritage warts by renaming some functions and
  > eliminating a one line called once function by inserting its code
  > inline.
  > No functional change.
  > From Brad. (krw@)

  ~ if_sk.c                               

  > Don't reinitialize card at every media change.
  > From Brad, tested by Brad, wilfried@ and Johan M:son Lindman.
  > ok wilfried@ (krw@)

  ~ if_bge.c                              

  > Workarounds for DMA HW errata on pci express chips. Based on FreeBSD
  > changes but with the more conservative (i.e. specific) chip matching
  > logic of Linux's tg3. No change except to pci express chips.
  > From Brad.
  > Tested by brad & thib@ (BCM5750) and pedro la peu (BCM5752). (krw@)

  ~ if_bnx.c                              

  > bnx_init() takes a pointer to sc, not ifp. (mcbride@)

  ~ if_bnx.c                              ~ if_bnxreg.h

  > move the interrupt establishment till after everything in the softc is
  > set up and allocated (which happens in a mountroothook). this prevents an
  > early call to the interrupt handler from causing a null deref when trying
  > to look into the unallocated regions.
  > found by mcbride when ciss and bnx were sharing an interrupt. mounting
  > root caused interrupts before the bnx was properly set up.
  > "commit your fix" mcbride@ (dlg@)

  ~ if_bnx.c                              

  > Remove bogus check for old firmware.
  > Identical fixes from myself and brad@, also reported by chefren at pi.net.  > (mcbride@)

  ~ if_em.c                               

  > Fix typo.  From brad@ (kettenis@)

dev/usb

  ~ usbdevs                               ~ if_zyd.c

  > Match some additional devices found in Linux driver. (jsg@)

  ~ usbdevs.h                             ~ usbdevs_data.h

  > regen (jsg@)

  ~ usbdevs                               

  > Correct entry for D-Link DUB-E100 rev B1 (jsg@)

  ~ usbdevs.h                             ~ usbdevs_data.h

  > regen (jsg@)

  ~ if_axe.c                              ~ if_axereg.h

  > Add support for AX88772 devices with external PHY.
  > Thanks to Marco S Hyman for sending me hardware to play with.
  > Committed over DUB-E100 rev B1 with external IC Plus PHY. (jsg@)

  ~ usbdevs.h                             ~ usbdevs_data.h

  > regen (jsg@)

  ~ usbdevs                               ~ if_axe.c

  > Add a bunch of as yet untested AX88178 devices. (jsg@)

isofs/cd9660

  ~ cd9660_vnops.c                        

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

kern

  ~ kern_acct.c                           ~ kern_sig.c
  ~ uipc_usrreq.c                         ~ vfs_syscalls.c
  ~ vfs_vnops.c                           ~ vfs_default.c
  ~ vnode_if.src                          

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

  ~ vnode_if.c                            

  > regen (thib@)

  ~ kern_fork.c                           

  > reintroduce rev 1.83 -- put the new process on the proc lists
  > later when more of it's resources have been allocated and thus
  > kill(2)ing such a process has more predictable results.
  > now w/ a couple of kettenis remarks; kettenis@ miod@ ok (mickey@)

  ~ kern_sig.c                            

  > When sending signals, instead of checking if the process sending the
  > signal to is the curproc, check if it's SONPROC. That way we'll also
  > signotify processes on other CPUs. This gives MD code the opportunity
  > to actually deliver signals to other CPUs faster than the clock
  > interrupt (in the worst case).
  > This is a NOP until MD code is fixed.
  > miod@ ok (art@)

  ~ uipc_domain.c                         

  > allow kernels with TCP_SIGNATURE (aka tcp md5sig), but without IPSEC to
  > compile and work. need to register pfkey whenever tcp md5 or ipsec is
  > defined, and the various ipsec encapsulations only if ipsec is defined.
  > ok theo (henning@)

miscfs/fifofs

  ~ fifo.h                                ~ fifo_vnops.c

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

miscfs/specfs

  ~ spec_vnops.c                          ~ specdev.h

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

msdosfs

  ~ msdosfs_vnops.c                       ~ denode.h

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

net

  ~ pfkeyv2.c                             

  > allow kernels with TCP_SIGNATURE (aka tcp md5sig), but without IPSEC to
  > compile and work. need to register pfkey whenever tcp md5 or ipsec is
  > defined, and the various ipsec encapsulations only if ipsec is defined.
  > ok theo (henning@)

netinet

  ~ ip_ipsp.c                             

  > allow kernels with TCP_SIGNATURE (aka tcp md5sig), but without IPSEC to
  > compile and work. need to register pfkey whenever tcp md5 or ipsec is
  > defined, and the various ipsec encapsulations only if ipsec is defined.
  > ok theo (henning@)

netinet6

  ~ icmp6.c                               

  > fix infinite loop in case nip6 and nicmp6 are not on the same mbuf.  NetBSD
  > PR 34994+35333 (itojun@)

  ~ icmp6.c                               

  TAGGED OPENBSD_4_0
  > Errata #8:
  > Under some circumstances, processing an ICMP6 echo request would cause
  > the kernel to enter an infinite loop. (miod@)

  ~ icmp6.c                               

  TAGGED OPENBSD_3_9
  > Errata #18:
  > Under some circumstances, processing an ICMP6 echo request would cause
  > the kernel to enter an infinite loop. (miod@)

  ~ icmp6.c                               

  TAGGED OPENBSD_3_9
  > oops, previous commit was incorrect. (itojun@)

  ~ icmp6.c                               

  TAGGED OPENBSD_4_0
  > Update to errata #8, merge revision 1.92 as well. (miod@)

  ~ icmp6.c                               

  TAGGED OPENBSD_3_9
  > Update to errata #18, merge revision 1.92 as well. (miod@)

nfs

  ~ nfs.h                                 ~ nfs_vnops.c
  ~ nfsnode.h                             

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

  ~ nfs_vnops.c                           

  > don't assume our directory vnode will be flushed out after us, fixes
  > crash when unmounting nfs volumes, okay thib@, krw@ and mickey@ (pedro@)

scsi

  ~ scsi_ioctl.c                          

  > Report SDEV_UMASS devices as non-'SCSI', i.e. 'ATAPI', devices when
  > asked via SCIOCIDENTIFY. More closely emulates the Linux
  > SCIOCIDENTIFY/SG_EMULATED_HOST combination used by cdparanoia (and
  > god knows what other ports). Nothing in the tree itself uses
  > SCIOCIDENTIFY.
  > Fixes cdparanoia's handling of USB CD drives that can't handle the
  > MODE SENSE command. cdparanoia sends 'ATAPI' devices only MODE SENSE
  > BIG (0x5a) commands. Finding actual SCSI devices that might have
  > trouble with MODE SENSE BIG (i.e. very old devices) behind USB is
  > deemed highly unlikely.
  > Problem reported, debugged and fix tested by bernd at .  > ok marco@ millert@ miod@ (krw@)

  ~ st.c                                  

  > Add media_fileno, media_blkno into st_softc and report their values
  > back via MTIOCGET ioctl, rather than always reporting -1 (unknown).
  > Simple 'Set to 0 on rewind, increment on file marks and positioning,
  > and set back to -1 on any ambiguity' logic. No functional impact on
  > current operation.
  > Should allow things like Bacula to work much faster by keeping track
  > of the current media position.
  > Adapted from FreeBSD code.
  > ok beck@ (krw@)

sys

  ~ vnode.h                               

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

  ~ vnode_if.h                            

  > regen (thib@)

ufs/ext2fs

  ~ ext2fs_vnops.c                        

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

ufs/ffs

  ~ ffs_balloc.c                          

  > Fix growing of direct blocks in ffs2_balloc() (pedro@)

  ~ ffs_softdep.c                         

  > UFS2 -> FFS2 (pedro@)

  ~ ffs_softdep.c                         

  > Fix splbio() in initiate_write_inodeblock_ufs2() (pedro@)

  ~ ffs_vnops.c                           

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

  ~ ffs_balloc.c                          

  > correctly initialise unwindidx in ffs2_balloc()
  > ok pedro (sturm@)

  ~ ffs_softdep.c                         

  > more ufs2 leftovers (pedro@)

  ~ ffs_balloc.c                          

  > "correctly deallocate dependencies when growing into an indirect block
  > and failing to acquire a data block, from freebsd"
  > fix this in ffs2_balloc() as well
  > requested by and ok pedro (sturm@)

ufs/ufs

  ~ ufs_extern.h                          

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

uvm

  ~ uvm_pglist.c                          

  > Implement multiple segment allocation for uvm_pglistalloc, which fixes
  > most agp_generic_bind_memory failures when memory is limited and very
  > fragmented.
  > In effect, this should fix a lot of X startup crashes after
  > activities that exercise memory a lot (e.g. make builds, building big
  > ports, etc).
  > ok mickey, miod (dim@)

xfs

  ~ xfs_vopdefs.h                         

  > Retire VOP_LEASE(); It was a bit for NQNFS and has
  > effectively been a no-op for quite some time now,
  > without promise for future usage.
  > ok pedro@
  > Testing by krw@ (earlier diff)
  > and Johan Mson Lindman (tybollt at solace.miun.sethib@)

== usr.bin =========================================================== 08/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin

chpass

  ~ chpass.c                              

  > Fix -a when given an entry with an already existing user. Also, give
  > error message if a user arg has been given with -a. Noted by Dan
  > Brosemer. ok millert@ jaredy@ (otto@)

compress

  ~ main.c                                

  > We don't want to use GZIP for compress.  Only examine GZIP environment
  > variable when invoked as gzip, gunzip, gzcat, etc.	OK otto@ (millert@)

cvs

  ~ checkout.c                            

  > replace checkout_write_revision() with rcs_rev_write_stmp()
  > they do exactly the same anyway. (joris@)

  ~ checkout.c                            

  > reset sticky tags with -A; OK joris at .sa@)

  ~ commit.c                              ~ diff.c
  ~ file.c                                ~ import.c
  ~ rcs.c                                 ~ rcs.h

  > move things around in rcs_rev_getbuf() and rcs_rev_write_fd()
  > so that we do keyword expansion on-the-fly if required instead
  > of obtaining the revision in memory first, running over the revision
  > lines while expanding keywords and only then writing them to the fd
  > or memory buffer.
  > this drasticly decreases cpu usage by opencvs on very large trees (like
  > src).
  > OK niallo@ (joris@)

  ~ diff.c                                

  > replace old diffing mechanisms with the new fluffy RCS stuff,
  > okay niallo@ (joris@)

  ~ cvs.1                                 

  > document `update -C'. (xsa@)

  ~ rcs.c                                 

  > sanitize what kind of symbols we can have in symbol names,
  > tested & found brad@ (joris@)

  ~ checkout.c                            

  > handle -l and -R for checkout too. (xsa@)

  ~ checkout.c                            

  > better, share -l and -R flags between checkout and export cmds. (xsa@)

  ~ client.c                              

  > handle the export command in cvs_client_{set,clear}_sticky(). (xsa@)

  ~ checkout.c                            ~ client.c
  ~ cvs.h                                 ~ remote.h
  ~ server.c                              ~ update.c
  ~ util.c                                

  > add support for remote checkout in both our client and server,
  > with this the last real dependency to actually be able to use opencvs
  > in a remote setup has been added.
  > based upon an initial diff from xsa@
  > 'kay niallo@ (joris@)

  ~ checkout.c                            

  > send -l Argument to server when appropriate. (xsa@)

  ~ checkout.c                            

  > on checkout, only create the module path if we find the module.
  > Also make error ouput match GNU CVS.
  > ok joris@ (thib@)

  ~ checkout.c                            ~ cvs.h

  > put functions protos at the right place. (xsa@)

  ~ update.c                              

  > more informative error msg. (xsa@)

  ~ commit.c                              

  > send -l Argument to server when appropriate. (xsa@)

  ~ client.c                              ~ import.c
  ~ remote.c                              ~ remote.h
  ~ server.c                              

  > import improvements:
  > - add support for remote import in opencvs client and server.
  > - do not free already free'd buffers in import_update()
  > - do not append a '\0' to the loaded file buffer, this was bad behaviour
  > makes import work fine both locally and remotely. (joris@)

  ~ root.c                                

  > extra safety check for NULL value.
  > ok joris@ (niallo@)

  ~ cvs.h                                 ~ import.c

  > move cvs_import() proto to the correct place. (xsa@)

  ~ file.c                                ~ repository.c

  > When we hit a DT_UNKNOWN for dirent->d_type in our filelist functions,
  > stat() the filepath to get the correct type, and therefor
  > add it to the appropriate list.
  > This fixes opencvs with CVSROOTs on NFS and AFS, because they
  > both return D_UNKNOWN for directories (afs even for files too)
  > in dirent->d_type.
  > problem found by thib@ and todd@
  > tested by thib@, todd@,
  > ok niallo@ (joris@)

  ~ util.c                                

  > another place to check for DT_UNKNOWN for d_type for afs/nfs (todd@)

  ~ file.c                                

  > check return value of rcs_translate_tag().	prevents a segfault found by
  > todd at .till need to fix tag translation though.
  > ok joris@ (niallo@)

  ~ diff.c                                ~ import.c
  ~ util.c                                

  > sweep opencvs for missing rcs_translate_tag() return checks. (niallo@)

  ~ file.c                                

  > more safety checks for file->file_rcsrev;
  > 'cool' niallo@ (joris@)

  ~ date.y                                

  > Prevent negative array index accesses when strlen(buff) == 0.
  > OK joris@ (ray@)

  ~ util.c                                

  > Add missing check for fgets return value in
  > cvs_get_repository_name().
  > OK joris@ (thib@)

  ~ file.c                                

  > typo (joris@)

  ~ admin.c                               

  > add support for the -A flag;
  > wrap some lines while there.
  > OK joris at .sa@)

lex

  ~ flex.skl                              ~ initscan.c

  > Eliminate gcc -Wundef warnings in flex-generated code.
  > ok miod@ (tsi@)

make

  ~ dir.c                                 ~ dir.h
  ~ parse.c                               

  > * add an extra argument to the dir lookup functions to say whether we
  > should look in the current directory.
  > * refactor include files treatment to have a separate function to look
  > for an include.
  > * use both pieces to make sure system includes do not get looked up
  > in the current directory unless everything else fails.
  > This is != from the netbsd solution to the same problem, which stuffs
  > `magic' entries into filepaths to say whether dot should be looked at
  > first or last.
  > feature requested by matthieu at .  > okay matthieu@ (espie@)

rcs

  ~ ci.c                                  

  > When checking in multiple files, if no log message is entered for
  > a file, don't prompt the user to reuse the (non-existent) log
  > message for the next.  Makes ci behave like the GNU version.
  > OK xsa@, joris@, niallo@ (millert@)

  ~ date.y                                

  > Prevent negative array index accesses when strlen(buff) == 0.
  > OK joris@ (ray@)

ssh

  ~ readconf.c                            

  > Honour activep for times (eg ServerAliveInterval) while parsing
  > ssh_config and ~/.ssh/config so they work properly with Host directives.
  > From mario.lorenz at wincor-nixdorf.coma bz #1275.	ok markus@ (dtucker@)

  ~ auth-skey.c                           ~ clientloop.c
  ~ kex.c                                 ~ session.c
  ~ ssh-keygen.c                          

  > spaces (stevesk@)

  ~ readconf.c                            

  > spaces (stevesk@)

yacc

  ~ PSD.doc/ss0                           

  > realively -> relatively
  > ``yep'' jmc@ (ray@)

== usr.sbin ========================================================== 09/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin

hoststated

  ~ check_tcp.c                           

  > when retrying a write, the position was previously not updated.
  > ok reyk@ (pyr@)

ntpd

  ~ sensors.c                             

  > Add some comments on the expresssion which converts sensor timedeltas
  > to ntp offsets; also, rewrite the expression to make it more clear
  > with no change in semantics.  ok henning@ (otto@)

  ~ ntp.c                                 

  > Esape from the Mouth of Madness by adjusting stored sensor offsets
  > when we adjust time. This prevents ntpd from going wild when using
  > sensor time sources; ok henning@ (on an earlier version) and a LOT
  > of testing by naddy@ (otto@)

  ~ ntp.c                                 ~ ntpd.c
  ~ ntpd.h                                ~ sensors.c
  ~ server.c                              

  > Although Unix compilers accept more than one definition of a global
  > symbol, follow the guidelines from K&R: only one definition of a
  > global symbol (and possibly more declarations).  Rename some vars
  > here and there to avoid shadowing. ok henning@ (otto@)

  ~ sensors.c                             

  > the new sensors tre can give us the number of sensors per type.
  > With this patch, we give up without bothering sysctl
  > kern_sensors.c::sensor_find() unless we know for sure that timedelta
  > sensor is present.
  > From: "Constantine A. Murenin" <cnst+openbsd at bugmail.mojo.ruhenning@)

ospfd

  ~ parse.y                               

  > typo
  > metricr -> metric (michele@)

  ~ ospfd.c                               ~ rde.c

  > Ignore SIGPIPE in the parent and RDE as well. (claudio@)

ripd

  ~ parse.y                               ~ rip.h

  > allow users to specify interface specific parameters even on a global basis
  > OK henning@ (michele@)

vnconfig

  ~ vnconfig.8                            

  > clarify that -K needs an argument >= 1000, and add two examples dealing
  > with
  > encrypted images.
  > help from and ok pedro@ jmc@ (grunk@)

===============================================================================



More information about the owc mailing list